Issues with a join on the summary_view dataset: "Cannot publish dataset Make sure your workflow is valid with one single last step as the output." error

I’m attempting to join a .csv file to the summary_view dataset. This is my first time attempting a join with the newer, more visual dataset view.

When attempting this left join on summary_view, I receive the following error: "Cannot publish dataset

Make sure your workflow is valid with one single last step as the output."

Long term I’ve planned to move the .csv data into Athena, I’m just trying to get this to work for now. I was able to successfully create this join in the past.

Here’s an illustration of the dataset, where the .csv file and related join are the only items added to the default config.

While not perhaps the most efficient join, this seems fairly straightforward to me. Thanks for any guidance in this area!

Hi @jkrewquicksight ,

Can you explain a bit on why are you structuring your joins this way? The error “Cannot publish dataset. Make sure your workflow is valid with one single last step as the output” means your workflow has multiple terminal (end) nodes instead of converging into a single final output. In the new data prep experience, the join workflow must form a linear chain, each join’s result feeds into the next, and the entire graph terminates at exactly one point. Based on your diagram, the CSV file, appears to be connected in a way that creates a branch, and that essentially gives your workflow two endpoints. To fix this, you will have to either restructure your joins or add a join or append step to recombine the paths back into a single output node.

While I do not understand the full context of your data, a potential restructuring you can do is:
summary_view → LEFT JOIN → account_map → LEFT JOIN → ri_sp_mapping → RIGHT JOIN-> CSV

Thank you for your thoughtful reply. Regarding the use case, I’m layering some ownership mapping on top of the data, joining on account id.

I had made an assumption that I needed to join directly against the core dataset, “summary_view”, vs being able to add a join along the end of the existing chain.

I took your advice and added the joined data at the end of the chain (assuming I’m using the correct terminology-I don’t recall ever hearing this term when using SQL. I would probably call this multiple joins.)

I believe this is working as expected:

My perspective is that I’m trying to add various mappings without messing up any of the default configs created via the CID templates.

Thanks again for your response, and if my approach illustrated in this post looks suspect, let me know.

Hi @jkrewquicksight ,

Glad this is working as expected! Honestly, in other cases I would have also called this multiple joins but I stated chains in this case as once you added all the joins it visually looks similar to a chain. Overall, nothing is standing out on my end! Please let me know if you have any additional questions related to your initial post.

Thank you

Hi @jkrewquicksight,

Just checking back in since we haven’t heard from you in a bit. I wanted to see if the guidance shared earlier helped resolve your question, or if you found a solution in the meantime.

If you still have any additional questions related to your initial post, feel free to share them. Otherwise, any update you’re able to provide within the next 3 business days would be helpful for the community.

Thank you

Hi @jkrewquicksight ,

Since I haven’t received any further updates from you, I’ll treat this inquiry as complete for now. If you have any additional questions, feel free to create a new post in the community and link this discussion for context.

Thank you